I was at UCSF when Prusiner was doing some of his best work on prions- it typically involved infecting mice and waiting 2 months to get a result that could be analyzed molecularly.
Anyway, this model system: https://en.wikipedia.org/wiki/Fungal_prion is pretty well understood, and it's generally hoped that understanding its molecular behavior will go a long way to understanding mammalian prions.
I think the effort and the debate are worth it though because on the top of it all, prion or not, the root cause of Parkinson's disease is still not known. At least several million bucks of taxpayer money will probably go to prion-related grants because of the publicity and interested of various research groups, and while the prion idea may pan out or not, it can only add to the effort to figure out how this all starts in the body.
Did you mean s/be coding molecules/have catalytic function/ ?
I'm referring to the determination that DNA, not protein, was the hereditary molecule. I probably just worded this poorly, to give the implication that coding capability was limited to proteins. It's just that most people at the time thought that DNA was found in either nA (a string of As), nT, nG, or nC, or random arrangements. More than anything people just didn't think that DNA could "code" for anything, while proteins could, because people didn't seem to realize that unique sequences of heteropolymers was a valid form of coding. It was mostly thought that the information was stored in huge structures formed by proteins, and that the coding information was architectural.
See Mayr, "Origin of Biological Thought", final chapter, starting with the section on Meischer. Or Crick's "On Protein Synthesis" where he proposed the linear code: http://profiles.nlm.nih.gov/ps/access/scbbzy.pdf